Skip to content
CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing
Branch: master
Clone or download
Latest commit 3c542e2 Jul 16, 2018
Type Name Latest commit message Commit time
Failed to load latest commit information.
dataset dataset agreement Jul 16, 2018
documentation agree Jul 16, 2018
src refactor Jul 16, 2018 contributors Jul 16, 2018

Alt text


Contributors: Jie Zhang, Zhanyong Tang, Meng Li, Dingyi Fang, Petteri Nurmi, Zheng Wang

Northwest University, China, Lancaster University, UK, University of Helsinki, Finland

CrossSense is an open source framework for scaling up WiFi sensing to new environments and larger problems. It uses machine learning techniques to address the problem. To reduce the cost of sensing model training data collection, CrossSense employs machine learning to train, off-line, a roaming model to generate, from one set of measurements, synthetic training samples for each target environment. To scale up to a larger problem size, CrossSense adopts a mixture-of-experts approach where multiple specialized sensing models, or experts, are used to capture the mapping from diverse WiFi inputs to the desired outputs.


CrossSense is built upon the python scikit-learn machine learning package.


To obtain our dataset, please follow the instructions here.


CrossSense is not production ready. It's a research prototype that demonstrates the viability of applying machine learning to scale up wireless based sensing. If you encounter any problems, please file an issue on github.


Source code of CrossSense is released under the Apache License (v2.0).


  title={CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing},
  author={Zhang, Jie and Tang, Zhanyong and Li, Meng and Fang, Dingyi and Nurmi, Petteri and Wang, Zheng},
  booktitle={The 24th ACM International Conference on Mobile Computing and Networking},
  series = {MobiCom '18},
You can’t perform that action at this time.